Use of Paper Medical Records Versus Administrative Data for Measuring and Improving Health Care Quality: Are We Still Searching for a Gold Standard?

Abstract

Medical record data has been the "gold standard" for calculating performance measures; however, the relative efficiency of using claims data has led to a rapid increase in the use of these and other types of administrative data. This article will explore the strengths and weaknesses of using existing sources of data for measuring quality in the ambulatory setting, and describe the results of a recent quality improvement initiative that measured quality based on both administrative data and medical record review. Data were obtained from a managed care organization's 1998 paid claims for 246 adult diabetic patients and 3441 female patients. Medical record data was abstracted from a sample of 924 adult diabetic patients and 598 female patients. All patients were seen at one of 15 primary care clinics during the study period. Performance indicators selected included two Health Plan Employer Data and Information Set (HEDIS) measures for women's health screening and three American Diabetes Association measures for diabetic care. Administrative data captured a greater proportion of patients receiving mammograms (66 vs. 46%), Pap tests (63 vs. 35%), retinal eye exams (44 vs. 27%), diabetics with hypertension or proteinuria on angiotensin-converting enzyme (ACE) inhibitors (53 vs. 41%), and diabetics with coronary artery disease on lipid lowering agents (79 vs. 52%). This study found that quality measures that could be calculated using administrative data showed higher rates of performance than indicated by a review of the medical record alone. Claims data is more accurate for identifying services with a high likelihood of documentation due to reimbursement and in situations where services are provided outside of the office.